1.Field of Technology
This invention relates to the field of wireless mobile communication and in particular to cellular communication. It is further related with HRPD, EV-DO, CDMA2000, session negotiation, dormant handoff, and the like. More particularly, the present invention relates to a method for triggering re-negotiation of a session when an Access Terminal moves from one access network (i.e. source AN) to another network (i.e. target AN) having different capabilities in a high rate packet data system.
2.Description of the Related Art
In CDMA 2000 1× High Rate Packet Data (HRPD) systems as defined by 3GPP2,a region is divided into subnets and subnets may contain multiple cells with each cell being served by a Base Transceiver Station (BTS). A cell is a physical grouping of one or more sectors that transmit the same power control command to an access terminal where each sector is a part of the access network that is identified as providing one CDMA channel. The Access Network (AN), in other words the network equipment providing data connectivity between a packet switched data network (typically the Internet) and the access terminal, covers one or more subnets. An Access Terminal (AT) is an entity, which a user can use to get a service from an AN in an HRPD system. An AT may be connected to a computing device such as laptop or personal computer. It can also be a self-contained data device such as a personal digital assistant. An Access Network (AN) in an HRPD system performs session negotiation, resource allocation and mobility management among other functions.
In an HRPD network, each AT is assigned a Unicast Access Terminal Identifier (UATI) which is universally unique and is used to address the mobile terminal when an AN sends messages to the AT. Whenever an AT crosses a subnet (e.g.: foot print of an AN), the AT sends a UATI Request message.
FIG. 1 shows the structure of a Sector ID/UATI and derivation of a subnet address from a Sector IDIUATI. The following can be observed:
The Sector Id is divided into two portions. The ‘n’ most significant bits (MSBs) represent the identifier for the subnet and the lower ‘128-n’ bits identify a particular sector within a subnet.
A subnet mask of length ‘n’ is a 128 bit value whose binary representation consists of ‘n’ consecutive ‘1’s followed by ‘128-n’ consecutive ‘0’s.
The subnet address from a Sector ID/UATI is obtained by performing a logical AND of the address and the subnet mask.
Each sector advertises its Sector ID, which is also a 128-bit address. This is how the AT knows that it has entered the foot-print of a new subnet.
A UATI has a similar structure as a Sector ID. A full UATI is of size 128 bits. A UATI is subdivided into UATI 104 and UATI 024, which signify 104 MSBs and 24 least significant bits (LSBs) of the UATI respectively. If an AN does not send the UATI 104 and UATISubnetMask fields in a UATIAssignment message, then they can be derived from SectorID [127: 24] and subnet mask of a SectorParameters message of that particular sector.
An AT sends a UATI request message on an Access Channel, which is the portion of the Reverse Channel that is used by ATs to communicate with the AN when they do not have a traffic channel assigned (there is a separate Reverse Access Channel for each sector of the AN).
While sending a UATI request message, an AT includes the “ColorCode|UATI[23:0]” in the MAC layer header of every Access Channel capsule, where ‘|’ denotes the concatenation operator. For Unicast packets, the MAC layer header of the Control Channel and Access Channel include the ‘ColorCode|UATI[23:0]’. Color code is used because sending a 128-bit UATI takes too much space in the Access and Control Channel messages. The subnet part of the sector ID is compressed to an 8-bit color code and is used as an alias for the subnet address. When the subnet of the sector changes, the color code changes. This is shown in FIG. 2.
ColorCode is only 8-bits and therefore cannot be globally unique. It implies that Color Codes can be reused. The ColorCode re-use scheme must ensure that there is no sector that has two or more neighboring sectors which are in different subnets but which use same ColorCode. An example ColorCode re-use scheme is shown in FIG. 3.
Once an AT determines that it has crossed a subnet (or AN foot-print), then the AT sends a UATI Request message to the new AN (which will be referred to as “target AN” henceforth) on access channel with UATIType=10 in Access Terminal Identifier record (ATI record) of MAC header. Upon seeing the UATI Request message, the target AN determines the address of the source AN.
The target AN may be provisioned with a table that maps the <Source ColorCode, Target SectorID> to the address of the source AN. In particular, for each target AN sector, a table may be provisioned that maps the ColorCode of each of the sector's adjacent subnets to the address of the AN responsible for the subnet. Hence the target AN determines the address of the source AN that corresponds to the ColorCode received in the MAC layer header by performing a table look-up in the appropriate above mentioned table.
An example of call flow for triggering the session retrieval and the actual session transfer from source AN to target AN is shown in FIG. 4. The following stages are involved in the process:                a. Once an AT recognizes that it crossed an AN (or subnet) footprint, the AT sends the UATI of an existing HRPD session (if available) to the target AN. The UATI can be used as an identifier for the existing HRPD session when the target AN attempts to retrieve the existing HRPD session State Information from the source AN.        b. The target AN sends an A13-Session Information Request message to the source AN to request the HRPD session information for the AT. The A13-Session Information Request message shall include the received UATI, the Security Layer Packet and Sector ID. The target AN starts timer TA13req.        c. The source AN validates the A13-Session Information Request and sends the requested HRPD session information of the AT to the target AN in an A13-Session Information Response message. The target AN stops timer TA13req.        d. The AT and the target AN complete the establishment of the HRPD session. Depending on the state of the AT and the target AN, either an existing HRPD session may be re-established, or a new HRPD session may be initiated if required. This step may be null if no further over-the-air signaling is required.        e. The target AN sends an A13-Session Information Confirm to the source AN to indicate that the target AN has received the HRPD session information. Upon receipt of the A13-Session Information Confirm message, the source AN deletes the associated AT HRPD session information.        
It is essential to mention here that the HRPD specification has various protocols each having many subtypes. Each protocol also has a default sub-type that is supported by all ATs and ANs, which are compliant to the HRPD specification. The HRPD specification has undergone a revision in which new subtypes for existing protocols and also new protocols and applications are introduced. By introducing the new protocol subtypes, new protocols and new applications, an AT's capabilities are enhanced and also Quality of Service (QoS) for many applications is ensured. For the purpose of discussion here, the first revision can be referred to as Rev-0 and the next revision as Rev-A.
Now, the case will be considered of an AT which is capable of Rev-A protocols but is powered up in a Rev-0 AN and hence the session negotiated uses all Rev-0 protocols. If the AT moves from the Rev-0 AN foot-print to a Rev-A AN foot-print, then the AT can recognize this movement by recognizing the change of subnet (by looking at the overhead messages being transmitted by the sector of the target AN). The following steps are executed after the AT recognizes the subnet boundary was crossed:                1.the AT sends a UATI Request message to Target AN.        2.the target AN retrieves this AT's session information from a source AN.        3.the AT and target AN continue using the session information and protocols that the target AN retrieved from the source AN.        
The major problem experienced here is that even though both the AT and the target AN are capable of supporting the advanced set of protocol subtypes, protocols and applications, they both continue using the default set of protocols as negotiated by the Rev-0 source AN. Hence, the AT and the target AN are not able to use efficient/optimized procedures and QoS mechanisms provided by the Rev-A HRPD specification. This is despite the fact that the Rev-A capable AT sends all the non-default subtypes for all protocol types that it supports (in the order of preference) in Configuration Request messages during the session negotiation with the Rev-1 HRPD target AN.
The drawback here is that new HRPD session is not being re-negotiated to upgrade the session protocols to Rev-A even if the Rev-A capable HRPD AT moves to Rev-A capable AN after setting up the initial session in a Rev-0 HRPD AN. This implies that efficient/optimized use of Rev-A protocol subtypes and improved QoS mechanisms are not possible even though both the AT and the AN are Rev-A capable.
Accordingly, there is a need for an improved apparatus and method for triggering session renegotiation between an access network and an access terminal used in a high rate packet data system.